#This script executes regressions to estimate
#associations between first-generation prenatal exposure to nonattainment 
#and first-generation assortative matching outcomes.

rm(list=ls())
gc()
library(reldist)
library(ineq)
library(stargazer)
library(data.table)
library(dplyr)
library(dtplyr)
library(ggplot2)
library(stringdist)
library(lfe)
library(haven)
library(readr)
library(broom)
library(tidylog)
library(rdrobust)

# Set the working directory
setwd("/projects/opp_env/caa1970/")

# Source external R scripts for custom functions
source("Code/Child_Outcomes/child_outcomes_functions.R")
source("Code/rounding.r")

# Load the first-generation assortative matching analysis dataset.
load("Data/first_gen_analysis_data_jepmicro_assortative.rda")


Table_B8 <- data.frame()
TabB8_A1 <- felm(partner_treated ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm |state_year+county_factor+year_factor+month+survey_year_by_year | (all_fullterm~nonattainment_infant)  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1, !is.na(partner_treated),!is.na(both_employed), !is.na(partner_higher_earnings), real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_A1)
TabB8_A1$N
temp_n<-  rounding_rules_counts(TabB8_A1$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44,married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm), real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$partner_treated, na.rm=T ),4)
temp_f <-  signif(TabB8_A1$stage1$iv1fstat[[1]]["F"],4)
temp_results <- tidy(TabB8_A1) %>% filter(term == "`all_fullterm(fit)`" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, fstat = temp_f,  table = "B4", column ="A1" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_A2 <- felm(both_employed ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm |state_year+county_factor+year_factor+month+survey_year_by_year | (all_fullterm~nonattainment_infant)  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44,married == 1,!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_A2)
TabB8_A2$N
temp_n<-  rounding_rules_counts(TabB8_A2$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm), real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$both_employed, na.rm=T ),4)
temp_f <-  signif(TabB8_A2$stage1$iv1fstat[[1]]["F"],4)
temp_results <- tidy(TabB8_A2) %>% filter(term == "`all_fullterm(fit)`" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, fstat = temp_f,  table = "B4", column ="A2" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_A3 <- felm(both_college ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm |state_year+county_factor+year_factor+month+survey_year_by_year | (all_fullterm~nonattainment_infant)  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_A3)
TabB8_A3$N
temp_n<-  rounding_rules_counts(TabB8_A3$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$both_college, na.rm=T ),4)
temp_f <-  signif(TabB8_A3$stage1$iv1fstat[[1]]["F"],4)
temp_results <- tidy(TabB8_A3) %>% filter(term == "`all_fullterm(fit)`" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, fstat = temp_f,  table = "B4", column ="A3" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_A4 <- felm(partner_higher_earnings ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm |state_year+county_factor+year_factor+month+survey_year_by_year | (all_fullterm~nonattainment_infant)  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1, !is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_A4)
TabB8_A4$N
temp_n<-  rounding_rules_counts(TabB8_A4$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1, !is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),!is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$same_earnings, na.rm=T ),4)
temp_f <-  signif(TabB8_A4$stage1$iv1fstat[[1]]["F"],4)
temp_results <- tidy(TabB8_A4) %>% filter(term == "`all_fullterm(fit)`" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, fstat = temp_f,  table = "B4", column ="A4" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_B1 <- felm(partner_treated ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm + nonattainment_infant|state_year+county_factor+year_factor+month+survey_year_by_year | 0 |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated),!is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_B1)
TabB8_B1$N
temp_n<-  rounding_rules_counts(TabB8_B1$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$partner_treated, na.rm=T ),4)
temp_results <- tidy(TabB8_B1) %>% filter(term == "nonattainment_infant" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl,  table = "B4", column ="B1" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_B2 <- felm(both_employed ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm + nonattainment_infant|state_year+county_factor+year_factor+month+survey_year_by_year | 0  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_B2)
TabB8_B2$N
temp_n<-  rounding_rules_counts(TabB8_B2$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$both_employed, na.rm=T ),4)
temp_results <- tidy(TabB8_B2) %>% filter(term == "nonattainment_infant" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, table = "B4", column ="B2" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_B3 <- felm(both_college ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm + nonattainment_infant|state_year+county_factor+year_factor+month+survey_year_by_year | 0  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44,married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_B3)
TabB8_B3$N
temp_n<-  rounding_rules_counts(TabB8_B3$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1,!is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings), !is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$both_college, na.rm=T ),4)
temp_results <- tidy(TabB8_B3) %>% filter(term == "nonattainment_infant" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, table = "B4", column ="B3" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset

TabB8_B4 <- felm(partner_higher_earnings ~factor(parent_race)+factor(parent_female)+PI_PC*year_trend+
                   PI_PC*I(year_trend^2) +epop*year + total_transfers*year_trend + 
                   county_pop*year_trend + epop*I(year_trend^2) + 
                   total_transfers*I(year_trend^2) + county_pop*I(year_trend^2) +
                   tmax_fullterm+ tmin_fullterm+prec_fullterm+anyprec_fullterm + nonattainment_infant|state_year+county_factor+year_factor+month+survey_year_by_year | 0  |county_factor, 
                 data=parent_piks_assortative%>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1, !is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),  real_wages<wage_cap, !is.na(parent_age_min)))

summary(TabB8_B4)
TabB8_B4$N
temp_n<-  rounding_rules_counts(TabB8_B4$N)
temp_ctrl <- signif(mean((parent_piks_assortative %>% filter(year%in%1969:1975, AGE >33 & AGE < 44, married == 1, !is.na(all_fullterm),!is.na(partner_treated), !is.na(both_employed), !is.na(partner_higher_earnings),!is.na(all_fullterm),  real_wages<wage_cap, !is.na(PI_PC),nonattainment_infant == 0))$same_earnings, na.rm=T ),4)
temp_results <- tidy(TabB8_B4) %>% filter(term == "nonattainment_infant" ) %>% #tidy regression, keep coef of interest
  mutate_each(funs(signif(.,4)), -term) %>% #round appropriately
  mutate(N=temp_n, ctrl_mean=temp_ctrl, table = "B4", column ="B4" ) #add variables for N and control mean, plus index the table number and column
Table_B8 <- bind_rows(Table_B8, temp_results) #Appended to big dataset
write_csv(Table_B8, "/projects/opp_env/caa1970/JPE_Micro/Output/Table B8/Table_B8.csv")